Policy-driven ai / ML client selection in a communications network

Abstract

This disclosure provides a method for selection of Artificial Intelligence / Machine Learning (AI / ML) Enablement Clients in a communications network. The method comprises transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating at the second network node a selection or reselection of AI / ML Enablement Clients; receiving at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the method further comprises determining at the second network node QoS degradation of an AI / ML Enablement Client.

Description

[0001] POLICY-DRIVEN AI / ML CLIENT SELECTION IN A COMMUNICATIONS NETWORK

[0002] TECHNICAL FIELD

[0003] The present invention generally relates to telecommunications networks, particularly 3GPP (3rd Generation Partnership Project) networks, and more specifically, the present invention relates to the selection and management of AI / ML (Artificial Intelligence / Machine Learning) Enablement Clients.

[0004] BACKGROUND

[0005] In 3GPP (3rd Generation Partnership Project) networks, the Service Enabler Architecture Layer (SEAL) is an architecture for enabling application services within telecommunications networks. The SEAL client(s) are integral elements situated on the user equipment (UE) side. These clients are engineered to facilitate direct interactions with the VAL (Vertical Application Layer) client(s). The SEAL servers provide the server-side functionalities corresponding to each service within the SEAL layer. Their primary role is to support and manage interactions with the VAL server(s). The servers include group management server, location management server, configuration management server, and others responsible for identity management, key management, and network resource management. The SEAL layer includes different services, e.g., SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0006] Artificial Intelligence / Machine Learning (AI / ML) technologies in 3GPP networks includes the integration of AI / ML processes, such as data collection, model training, and analytics provisioning. Federated Learning (FL) represents a specific application of AI / ML in telecommunications, allowing for the decentralized training of ML models across multiple devices or network nodes. This approach leverages data generated across the network by reducing the need to centralize AI / ML processes.

[0007] The integration of AI / ML technologies into 3GPP networks requires an efficient client selection mechanism for identifying and managing the nodes or devices that participate in AI / ML processes, since the selection of clients impacts the overall performance, scalability, and reliability of AI / ML-enabled services within the network.

[0008] A problematic aspect of the existing solutions concerns the selection and management of AI / ML Enablement Clients, particularly for AI / ML or FL operations. This leads to inefficiencies in how AI / ML or FL operations are initiated, managed, and scaled across the network, affecting the overall performance and reliability of AI / ML-enabled services.

[0009] A further problematic aspect of the existing solutions concerns dynamically adjusting client selection based on real-time network conditions and historical or predictive analytics. Thus, the AI / ML Enablement Clients may not operate optimally under varying network conditions, which can lead to degraded service quality and reduced effectiveness of AI / ML operations, especially in federated learning scenarios.

[0010] SUMMARY

[0011] The invention is set out in the appended set of claims.

[0012] An object of the invention is to enhance the selection and management of AI / ML Enablement Clients within 3GPP networks through a policy-driven approach, by introducing a mechanism for client selection and re-selection based on policies.

[0013] This disclosure provides a method for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating at the second network node a selection or reselection of AI / ML Enablement Clients; receiving at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the method further comprises determining at the second network node QoS degradation of an AI / ML Enablement Client. In some embodiments, the method further comprises initiating at the second network node adjusting of network QoS based on the at least one AI / ML policy. In some embodiments, the method further comprises initiating at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0014] An aspect of the invention relates to a method performed by a first network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; and receiving at the first network node from the second network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0015] An aspect of the invention relates to a method performed by a second network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises receiving at a second network node from a first network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating at the second network node a selection or reselection of AI / ML Enablement Clients; receiving at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the method further comprises determining at the second network node QoS degradation of an AI / ML Enablement Client. In some embodiments, the method further comprises initiating at the second network node adjusting of network QoS based on the at least one AI / ML policy. In some embodiments, the method further comprises initiating at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0016] Other aspects of the invention relate to mobile network nodes, particularly a first network node (116, 500), a second network node (117, 600), a third network node (118), a fourth network node (119) configured to perform the respective methods as described herein. Other aspects of the invention relate to computer program and computer program products.

[0017] In some embodiments, the first network node is a VAL Server (VAL Server). In some embodiments, the second network node is an AI / ML Enablement Server (AI / ML Enablement Server). In some embodiments, the third network node is an AI / ML Enablement Clients (AI / ML Enablement Clients). In some embodiments, the fourth network node is a NEF, NWDAF, or SEAL service (NEF, NWDAF).

[0018] Advantageously, the solution disclosed herein enables a dynamic and flexible approach to AI / ML client selection and management, allowing for tailored selections that match network conditions and requirements.

[0019] Advantageously, the solution disclosed herein introduces a policy-driven AI / ML Enablement client selection mechanism that operates based on predefined policies. This aspect ensures that client selection is governed by a set of defined policy parameters, enhancing effectiveness and predictability.

[0020] Advantageously, the solution disclosed herein also incorporates an advanced AI / ML Enablement client selection methodology that utilizes historical data and predictive services. By leveraging past performance and predictive analytics.

[0021] Advantageously, the solution disclosed herein further provides visibility into the status of the selected AI / ML client for the requestor. This feature ensures that the entities making requests are kept informed about the client's status.

[0022] Advantageously, the solution disclosed herein allows for real-time optimization of network performance, ensuring effectiveness and efficiency of AI / ML operations under fluctuating network conditions.

[0023] Additional objectives, features and advantages of the concepts disclosed herein will be apparent from the following description, claims and drawings, or may be learned by practice of the described technologies and concepts as set forth herein.

[0024] BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to best describe the manner in which the disclosed concepts may be implemented, as well as define other objects, advantages and features of the disclosure, a more particular description is provided below and is illustrated in the appended drawings. Understanding that these drawings depict only exemplary embodiments of the invention and are not therefore to be considered to be limiting in scope, the examples will be described and explained with additional specificity and detail through the use of the accompanying drawings.

[0026] Figure 1 illustrates an example networked system in accordance with particular embodiments of the solution described herein.

[0027] Figure 2 illustrates an example signaling diagram showing a procedure according to particular embodiments of the solution described herein.

[0028] Figure 3 illustrates an example flowchart showing a method performed by a mobile network node according to particular embodiments of the solution described herein. Figure 4 illustrates an example flowchart showing a method performed by a mobile network node according to particular embodiments of the solution described herein.

[0029] Figure 5 illustrates an example block diagram of a mobile network node configured in accordance with particular embodiments of the solution described herein.

[0030] Figure 6 illustrates an example block diagram of a mobile network node configured in accordance with particular embodiments of the solution described herein.

[0031] Figure 7 illustrates an example block diagram of a virtualized environment.

[0032] DETAILED DESCRIPTION

[0033] The invention will now be described in detail hereinafter with reference to the accompanying drawings, in which examples of embodiments or implementations of the invention are shown. The invention may, however, be embodied or implemented in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of present invention to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Components from one embodiment may be tacitly assumed to be present / used in another embodiment. These embodiments of the disclosed subject matter are presented as teaching examples and are not to be construed as limiting the scope of the disclosed subject matter. For example, certain details of the described embodiments may be modified, omitted, or expanded upon without departing from the scope of the described subject matter.

[0034] The example embodiments described herein arise in the context of a telecommunications network, including but not limited to a telecommunications network that conforms to and / or otherwise incorporates aspects of a fifth generation (5G) architecture. Figure 1 is an example networked system 100 in accordance with example embodiments of the present disclosure. Figure 1 specifically illustrates User Equipment (UE) 101 , which may be in communication with a (Radio) Access Network (RAN) 102 and Access and Mobility Management Function (AMF) 106 and User Plane Function (UPF) 103. The AMF 106 may, in turn, be in communication with core network services including Session Management Function (SMF) 107 and Policy Control Function (PCF) 111. The core network services may also be in communication with an Application Server / Application Function (AS / AF) 113. Other networked services also include Network Slice Selection Function (NSSF) 108, Authentication Server Function (ALISF) 105, User Data Management (UDM) 112, Network Exposure Function (NEF) 109, Network Repository Function (NRF) 110, Unified Data Repository (UDR) 114, Network Data Analytics Function (NWDAF) 115 and Data Network (DN) 104. In some example implementations of embodiments of the present disclosure, each one of the entities in the networked system 100 are considered to be a Network Function (NF). One or more additional instances of the NFs may be incorporated into the networked system.

[0035] The solution described herein aims to enhance the selection and management of AI / ML Enablement Clients within 3GPP networks through a policy-driven approach, by introducing a mechanism for client selection and re-selection based on policies.

[0036] This disclosure provides a method for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating at the second network node a selection or reselection of AI / ML Enablement Clients; receiving at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the method further comprises determining at the second network node QoS degradation of an AI / ML Enablement Client. In some embodiments, the method further comprises initiating at the second network node adjusting of network QoS based on the at least one AI / ML policy. In some embodiments, the method further comprises initiating at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0037] An aspect of the invention relates to a method performed by a first network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; and receiving at the first network node from the second network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0038] An aspect of the invention relates to a method performed by a second network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network. The method comprises receiving at a second network node from a first network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating at the second network node a selection or reselection of AI / ML Enablement Clients; receiving at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements. In some embodiments, the method further comprises determining at the second network node QoS degradation of an AI / ML Enablement Client. In some embodiments, the method further comprises initiating at the second network node adjusting of network QoS based on the at least one AI / ML policy. In some embodiments, the method further comprises initiating at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy. In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations. In some embodiments, the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy. In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary. In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status. In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0039] This disclosure also provides mobile network nodes, particularly a first network node (1 16, 500), and a second network node (117, 600), configured to perform the respective methods as described herein. In some embodiments, the first network node is a VAL Server 1 16. In some embodiments, the second network node is an AI / ML Enablement 1 17. In some embodiments, the third network node is an AI / ML Enablement Client 1 18. In some embodiments, the fourth network node is a NEF, NWDAF, or SEAL service 1 19.

[0040] This disclosure also provides the corresponding computer program and computer program products comprising code, for example in the form of a computer program, that when run on processing circuitry of the mobile network nodes causes the mobile network nodes to perform the disclosed methods.

[0041] Advantageously, the solution disclosed herein enables a dynamic and flexible approach to AI / ML client selection and management, allowing for tailored selections that match network conditions and requirements.

[0042] Advantageously, the solution disclosed herein introduces a policy-driven AI / ML Enablement client selection mechanism that operates based on predefined policies. This aspect ensures that client selection is governed by a set of defined policy parameters, enhancing effectiveness and predictability.

[0043] Advantageously, the solution disclosed herein also incorporates an advanced AI / ML Enablement client selection methodology that utilizes historical data and predictive services. By leveraging past performance and predictive analytics. Advantageously, the solution disclosed herein further provides visibility into the status of the selected AI / ML client for the requestor. This feature ensures that the entities making requests are kept informed about the client's status.

[0044] Advantageously, the solution disclosed herein allows for real-time optimization of network performance, ensuring effectiveness and efficiency of AI / ML operations under fluctuating network conditions.

[0045] The solution and the features comprised therein are further described in what follows.

[0046] • The VAL server may request the AI / ML Enablement Clients selection via providing the list of the discovered AI / ML Enablement Clients or providing the required number of clients that the AI / ML Enablement Server shall select for the VAL server. Also, for the FL scenario the VAL server should provide the list of endpoints at the VAL server for receiving AIML data from the selected AI / ML Clients.

[0047] • The AI / ML Enablement Clients selection may include the list of the provisioned AI / ML policies that shall be enforced by the AI / ML Enablement Server for example: o Given: The AIML Enablement Client was selected as a member for participate in the AIML operation(s) and experiences congestion in the network. Action: The AIML Enablement Server observes the QoS degradation of the AIML client. The AIML Enablement Server configures the network QoS for the AIML client (based on provided member participation policies by the VAL server) in order to keep the AIML Enablement Client as selected member for the AI / ML operation(s). If failed, the AIML Enablement Server initiates the re-selection procedure. Result: AIML Enablement Server maintains the continuity of the ALML process via generating or updating 5G network policies based on the provided AIML member selection policies.

[0048] • In order to improve the AI / ML Enablement Clients selection, the AI / ML Enablement Server may interact with NEF, NWDAF, and SEAL to retrieve the historical, current data and predictions for the relevant metrics. For example, if the prediction for the QoS values does not match the QoS member re-selection criteria in the provided policy, this client will not be selected.

[0049] • The AI / ML Enablement Server may use NEF and / or SEALDD service in order to configure the AI / ML traffic sessions between VAL server and AI / ML Enablement Client(s) with the given QoS and monitor the AI / ML traffic characteristics.

[0050] • The AIML server provides the notification to the VAL server about the selected AIML client’s status update like re-selected and replaced with a new client. Relevant aspects of the solution disclosed herein further include:

[0051] • The VAL server can enforce previously provisioned AIML policies that shall be enforced;

[0052] • The AI / ML Enablement Server can use both SEALDD and NEF services to configure the AI / ML traffic connections;

[0053] • The AI / ML Enablement Server can use NEF, NWDAF, and SEAL to retrieve the historical, current data and predictions for the relevant metrics in order to improve the client selection. For example, if the prediction for the QoS values match the QoS member re-selection criteria in the provided policy, this client will not be selected.

[0054] • The AIML server provides the notification to the VAL server about the selected AIML client’s status update like re-selected and replaced with a new client.

[0055] Hereinafter, drawings showing examples of embodiments of the solution are described in detail.

[0056] Figure 2 is a signaling diagram illustrating a procedure for selection of Artificial Intelligence / Machine Learning (AI / ML) Enablement Clients in a communications network. The procedure is performed by a first network node (116, 500), a second network node (1 17, 600), a third network node (118), and a fourth network node (1 19). In some embodiments, the first network node is a VAL Server 116. In some embodiments, the second network node is an AI / ML Enablement 117. In some embodiments, the third network node is an AI / ML Enablement Client 1 18. In some embodiments, the fourth network node is a NEF, NWDAF, or SEAL service 119.

[0057] Assumptions:

[0058] 1 . The proposed solution is based on client-server architecture for federated learning.

[0059] 2. The VAL server may discover and receive a list of AIML enablement clients that are suitable and have available data for a particular AIML operation.

[0060] 3. The discovery operation may find a list of AIML enablement clients.

[0061] 4. The VAL server may provision the required AI / ML policies and have the corresponding policy ID(s).

[0062] The description of the steps is the following: 1 . A VAL server sends a request to an AIML enablement server to select a list of AIML enablement clients that have been discovered to meet the requirements for AIML operations. The AIML enablement client selection request includes the requestor identifier, security credentials, VAL service identifier, a list of AIML client IDs for inclusion into an AIML set or number of required AIML clients to be selected and includes list of AI / ML policies identifiers, endpoint at the VAL sever for receiving ML model update from the AIML clients, and may include the notification endpoint for the selected AI / ML Enablement Client’s status update.

[0063] 2. The AIML enablement server validates the selection request. The AIML enablement server further performs authentication and authorization checks to determine if the requestor is able to create a selected AI / ML members set.

[0064] 3. If the requestor is authorized and the List of AI / ML client IDs IE was provided in step 1 , the AI / ML Enablement Server negotiates with the selected AI / ML Enablement Clients and creates a selected AI / ML Enablement Clients set with the AIML Enablement Clients that confirmed their selection as selected members and assigns an identifier for the set (e.g., VAL group ID).

[0065] 4. If the requestor is authorized and the Number of the required AIML clients IE was provided in step 1 , the AI / ML Enablement Server shall discover the AI / ML Enablement Clients based on provided list of AI / ML Policy IDs. The AI / ML Enablement Server may interact with NEF (e.g., uses the NEF procedures for the AFsessionWithQoS and the NEF procedures for the AnalyticsExposure and in particular the UE Communication Analytics and DN Performance Analytics), SEAL services (e.g., SS_NetworkResourceMonitoring, SS_LocationArealnfoRetrieval APIs), NWDAF and / or ADAE services to obtain the historical data, measurements, and / or predictions for relevant parameters (e.g., QoS, location information) in order to check that the selected members will match the AI / ML policies conditions. The AI / ML Enablement Server shall negotiate with the selected AI / ML Enablement Clients candidates and create a selected AI / ML Enablement Clients set with the AIML Enablement Clients that confirmed their participation as selected members and assign an identifier for the set (e.g., VAL group ID).

[0066] The member selection mechanism in step 4 is implementation specific.

[0067] In steps 3 and 4, the AI / ML Enablement Server provisions the endpoint at the VAL sever for receiving ML model update from the AIML Enablement Clients.

[0068] 5. If the List of AI / ML policy IDs information element was provided in step 1 : a. the AI / ML Enablement Server may determine the QoS parameters for the AIML traffic session between the requestor and the selected AI / ML Enablement Client(s) and configure the AI / ML traffic session(s) via SEALDD (Sdd_RegularTransmission API) or NEF services (AfSessionWithQoS API); and b. the AI / ML Enablement Server shall interact with the NEF and / or SEAL services (including SEALDD) to establish the associated monitoring subscriptions. The AI / ML Enablement Server determines the relevant subscription procedures and the parameters for these subscriptions based on the inputs received from the AI / ML policy IDs.

[0069] The AIML enabler server can reuse SEAL group management for any necessary group management.

[0070] 6. The AIML enablement server sends an AIML enablement client selection response that includes the status of the selection request and the assigned AIML set identifier.

[0071] 7. The AI / ML Enablement Server monitors the notifications of the established subscriptions in step 4 and based on the received notifications enforces the AI / ML policies which conditions are met, e.g.: a. if the conditions for the re-selection action within the member selection policy are met, the AI / ML Enablement Server shall initiate the re-selection procedure (steps 4-5) to select and configure a new AI / ML Enablement Client as a selected member. The AI / ML Enabler Server may configure and initiate the AI / ML data transfer from dis-selected member to the selected member. When the AI / ML data transfer is fishished, the AI / ML Enablement Server shall cancel the related monitoring subscriptions configured in step 5b and remove the enforced policies; and b. if conditions for the QoS adjustment policy are met, the AI / ML Enablement Server shall generate and enforce the QoS policy(-ies). The AI / ML Enablement Server shall continue the QoS monitoring for the relevant AI / ML Enablement Client(s). If the AI / ML Enablement Client QoS characteristics do not match the required QoS values, the AI / ML Enablement Server shall initiate the re-selection procedure as described in step 7a.

[0072] 8. If notification endpoint for the selected AI / ML client’s status update was provided in step 1 , the AI / ML Enablement Server notifies the VAL server about the selected AI / ML client’s status update, e.g., the AI / ML Enablement Client A is re-selected and replaced by AI / ML Enablement Client B. Table 1 shows the request sent by a VAL server to an AIML enablement server for the AIML enablement client selection procedure.

[0073] Table 1 : Request for AIML enablement client selection procedure Table 2 shows the response sent by the AIML enablement server to the VAL server for the AIML enablement client selection procedure.

[0074] Table 2: Response for AIML enablement client selection procedure

[0075] Hereinafter, flowcharts showing examples of embodiments of the solution are described in detail.

[0076] The embodiments correspond to methods performed by and involving a first network node (116, 500), a second network node (117, 600), a third network node (118), and a fourth network node (119). In some embodiments, the first network node is a VAL Server 116. In some embodiments, the second network node is an AI / ML Enablement 117. In some embodiments, the third network node is an AI / ML Enablement Client 118. In some embodiments, the fourth network node is a NEF, NWDAF, or SEAL service 119.

[0077] Figure 3 is a flowchart illustrating a method performed by the first network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network.

[0078] In step S-301 , the first network node transmits to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process.

[0079] In step S-302, the first network node receives from the second network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements.

[0080] In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations.

[0081] In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary.

[0082] In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status.

[0083] In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0084] Figure 4 is a flowchart illustrating a method performed by the second network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network.

[0085] In step S-401 , the second network node receives from a first network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process.

[0086] In step S-402, the second network node initiates a selection or reselection of AI / ML Enablement Clients.

[0087] In step S-403, the second network node receives from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions.

[0088] In step S-404, the second network node determines QoS degradation of an AI / ML Enablement Client.

[0089] In step S-405, the second network node initiates adjusting of network QoS based on the at least one AI / ML policy.

[0090] In step S-406, the second network node initiates AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy.

[0091] In step S-407, the second network node transmits to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements.

[0092] In some embodiments, the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations.

[0093] In some embodiments, the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy.

[0094] In some embodiments, the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary.

[0095] In some embodiments, the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status.

[0096] In some embodiments, the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function (NEF), a Network Data Analytics Function (NWDAF), or Service Enabler Architecture Layer (SEAL) service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery (SEALDD), Location Management, Network Resource Management, Group Management, and Configuration Management.

[0097] Figure 5 is a block diagram illustrating elements of a mobile network node 500 of a mobile communications network. In some embodiments, the mobile network node 500 is a VAL Server 116. As shown, the mobile network node may include network interface circuitry 501 (also referred to as a network interface) configured to provide communications with other nodes of the core network and / or the network. The mobile network node may also include a processing circuitry 502 (also referred to as a processor) coupled to the network interface circuitry, and memory circuitry 503 (also referred to as memory) coupled to the processing circuitry. The memory circuitry 503 may include computer readable program code that when executed by the processing circuitry 502 causes the processing circuitry to perform operations according to embodiments disclosed herein. According to other embodiments, processing circuitry 502 may be defined to include memory so that a separate memory circuitry is not required. As discussed herein, operations of the mobile network node may be performed by processing circuitry 502 and / or network interface circuitry 501 . For example, processing circuitry 502 may control network interface circuitry 501 to transmit communications through network interface circuitry 501 to one or more other network nodes and / or to receive communications through network interface circuitry from one or more other network nodes. Moreover, modules may be stored in memory 503, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 502, processing circuitry 502 performs respective operations (e.g., operations discussed below with respect to Example Embodiments relating to core network nodes).

[0098] Figure 6 is a block diagram illustrating elements of a mobile network node 600 of a mobile communications network. In some embodiments, the mobile network node 600 is an AI / ML Enablement Server 117. As shown, the mobile network node may include network interface circuitry 601 (also referred to as a network interface) configured to provide communications with other nodes of the core network and / or the network. The mobile network node may also include a processing circuitry 602 (also referred to as a processor) coupled to the network interface circuitry, and memory circuitry 603 (also referred to as memory) coupled to the processing circuitry. The memory circuitry 603 may include computer readable program code that when executed by the processing circuitry 602 causes the processing circuitry to perform operations according to embodiments disclosed herein. According to other embodiments, processing circuitry 602 may be defined to include memory so that a separate memory circuitry is not required. As discussed herein, operations of the mobile network node may be performed by processing circuitry 602 and / or network interface circuitry 601 . For example, processing circuitry 602 may control network interface circuitry 601 to transmit communications through network interface circuitry 601 to one or more other network nodes and / or to receive communications through network interface circuitry from one or more other network nodes. Moreover, modules may be stored in memory 603, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 602, processing circuitry 602 performs respective operations (e.g., operations discussed below with respect to Example Embodiments relating to core network nodes).

[0099] Figure 7 is a block diagram illustrating a virtualization environment 700 in which functions implemented by some embodiments may be virtualized. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments 700 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment 700 includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a Service Management and Orchestration Framework via an 0-2 interface.

[0100] Applications 702 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment Q400 to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0101] Hardware 704 includes processing circuitry, memory that stores software and / or instructions executable by hardware processing circuitry, and / or other hardware devices as described herein, such as a network interface, input / output interface, and so forth. Software may be executed by the processing circuitry to instantiate one or more virtualization layers 706 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 708a and 708b (one or more of which may be generally referred to as VMs 708), and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer 706 may present a virtual operating platform that appears like networking hardware to the VMs 708.

[0102] The VMs 708 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer 706. Different embodiments of the instance of a virtual appliance 702 may be implemented on one or more of VMs 708, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.

[0103] In the context of NFV, a VM 708 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs 708, and that part of hardware 704 that executes that VM, be it hardware dedicated to that VM and / or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handling specific network functions that run in one or more VMs 708 on top of the hardware 704 and corresponds to the application 702.

[0104] Hardware 704 may be implemented in a standalone network node with generic or specific components. Hardware 704 may implement some functions via virtualization. Alternatively, hardware 704 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration 710, which, among others, oversees lifecycle management of applications 702. In some embodiments, hardware 704 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station. In some embodiments, some signaling can be provided with the use of a control system 712 which may alternatively be used for communication between hardware nodes and radio units.

[0105] Embodiments within the scope of the present invention may also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such tangible computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or combination thereof) to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such connection is properly termed a computer- readable medium. Combinations of the above should also be included within the scope of the tangible computer-readable media.

[0106] Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Computer-executable instructions also include program modules that are executed by computers in standalone or network environments. Generally, program modules include routines, programs, objects, components, and data structures that perform particular tasks or implement particular abstract data types. Computer executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represent examples of corresponding acts for implementing the functions described in such steps.

[0107] Those of skill in the art will appreciate that other embodiments of the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, and the like. Embodiments may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination thereof) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0108] Communication at various stages of the described system can be performed through a local area network, a token ring network, the Internet, a corporate intranet, 802.11 series wireless signals, fiber-optic network, radio or microwave transmission, etc. Although the underlying communication technology may change, the fundamental principles described herein are still applicable.

[0109] The various embodiments described above are provided by way of illustration only and should not be construed to limit the invention. For example, the principles herein may be applied to any remotely controlled device. Further, those of skill in the art will recognize that communication between the remote the remotely controlled device need not be limited to communication over a local area network but can include communication over infrared channels, Bluetooth or any other suitable communication interface. Those skilled in the art will readily recognize various modifications and changes that may be made to the present invention without following the example embodiments and applications illustrated and described herein, and without departing from the scope of the present disclosure.

[0110] The terminology used herein is for the purpose of describing various embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "includes," "including," "comprises," and "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, or components, and combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, or components, and combinations thereof. Further, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to ""a / an / the element, apparatus, component, means, module, step, etc."" are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, module, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

Claims

CLAIMS1 . A method for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network, the method comprising: transmitting (S-301) from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating (S-402) at the second network node a selection or reselection of AI / ML Enablement Clients; receiving (S-403) at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting (S-407) from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements.

2. The method of claim 1 , further comprising: determining (S-404) at the second network node QoS degradation of an AI / ML Enablement Client.

3. The method of any one of claims from claim 1 to claim 2, further comprising: initiating (S-405) at the second network node adjusting of network QoS based on the at least one AI / ML policy.

4. The method of any one of claims from claim 1 to claim 3, further comprising: initiating (S-406) at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy.

5. The method of any one of claims from claim 1 to claim 4, wherein the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations.

6. The method of any one of claims from claim 1 to claim 5, wherein the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy.

7. The method of any one of claims from claim 1 to claim 6, wherein the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary.

8. The method of any one of claims from claim 1 to claim 7, wherein the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status.

9. The method of any one of claims from claim 1 to claim 8, wherein the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function, NEF, a Network Data Analytics Function, NWDAF, or Service Enabler Architecture Layer, SEAL, service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery, SEALDD, Location Management, Network Resource Management, Group Management, and Configuration Management.

10. A method performed by a first network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network, the method comprising: transmitting from a first network node to a second network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; and receiving at the first network node from the second network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements.- T1 -11 . The method of claim 10, wherein the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations.

12. The method of any one of claims from claim 10 to claim 11 , wherein the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary.

13. The method of any one of claims from claim 10 to claim 12, wherein the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status.

14. The method of any one of claims from claim 10 to claim 13, wherein the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a Network Exposure Function, NEF, a Network Data Analytics Function, NWDAF, or Service Enabler Architecture Layer, SEAL, service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery, SEALDD, Location Management, Network Resource Management, Group Management, and Configuration Management.

15. A method performed by a second network node for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network, the method comprising: receiving (S-401) at a second network node from a first network node a selection request of AI / ML Enablement Clients based on at least one of: an indication of at least one client, an indication of a required number of clients, or an indication of at least one AI / ML policy, particularly wherein the policy is to be enforced during the client selection process; initiating (S-402) at the second network node a selection or reselection of AI / ML Enablement Clients; receiving (S-403) at the second network node from a fourth network node information or criteria relevant for client selection and participation, particularly wherein the information comprises any one of historical data, predictive data, historical network performance data or future network conditions predictions; and transmitting (S-407) from the second network node to the first network node an indication of the status updates of selected AI / ML Enablement Clients, particularly wherein the status updates include re-selections and replacements.

16. The method of claim 15, further comprising: determining (S-404) at the second network node QoS degradation of an AI / ML Enablement Client.

17. The method of any one of claims from claim 15 to claim 16, further comprising: initiating (S-405) at the second network node adjusting of network QoS based on the at least one AI / ML policy.

18. The method of any one of claims from claim 15 to claim 17, further comprising: initiating (S-406) at the second network node AI / ML traffic sessions between the VAL Server and AI / ML Enablement Clients with specified QoS parameters, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy, particularly wherein the initiating comprises setting dedicated network resources and / or QoS parameters for AI / ML data transmission based on the at least one AI / ML policy.

19. The method of any one of claims from claim 15 to claim 18, wherein the at least one AI / ML policy comprises at least one rule for client participation in AI / ML and / or federated learning operations.

20. The method of any one of claims from claim 15 to claim 19, wherein the method further comprises monitoring at the AI / ML Enablement Server real-time network traffic data to determine the QoS degradation and / or the AI / ML Enablement Client location, particularly wherein the monitoring comprises monitoring movement of the AI / MO Enablement Client in or out of a location defined in the at least one AI / ML policy.21 . The method of any one of claims from claim 15 to claim 20, wherein the selection or reselection of AI / ML Enablement Clients comprises applying at least one criteria for selection or reselection included in the at least one AI / ML policy, particularly wherein the criteria is for deciding when the selection or reselection of AI / ML Enablement Clients is necessary.

22. The method of any one of claims from claim 15 to claim 21 , wherein the indication of the status updates comprises information on the operational status of each AI / ML Enablement Client, particularly including any changes in their selection status.

23. The method of any one of claims from claim 15 to claim 22, wherein the first network node is a VAL Server, the second network node is a AI / ML Enablement Server, the third network node is at least one AI / ML Enablement Client, and the fourth node is a NetworkExposure Function, NEF, a Network Data Analytics Function, NWDAF, or Service Enabler Architecture Layer, SEAL, service, particularly wherein the SEAL service comprise any one of: SEAL Data Delivery, SEALDD, Location Management, Network Resource Management, Group Management, and Configuration Management.

24. Apparatus for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network, the apparatus comprising a processor and a memory, the memory containing instructions executable by the processor such that the apparatus is operable to perform the method of any one of claims from claim 10 to claim 14.

25. Apparatus for selection of Artificial Intelligence / Machine Learning, AI / ML, Enablement Clients in a communications network, the apparatus comprising a processor and a memory, the memory containing instructions executable by the processor such that the apparatus is operable to perform the method of any one of claims from claim 15 to claim 23.

26. A system comprising an apparatus as claimed in claim 24, and an apparatus as claimed in claim 25.

27. A computer-implemented system comprising one or more processors and one or more computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to perform a method according to any one of claims from claim 10 to claim 23.

28. A computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to perform a method according to any of claims from claim 10 to claim 23.

29. A computer program product, embodied on a non-transitory machine-readable medium, comprising instructions which are executable by a processor, causing the processor to perform the method according to any of claims from claim 10 to claim 23.

Images